white-dwarf-merger outcomes - welcome to the … · white-dwarf-merger outcomes simon jeffery...
TRANSCRIPT
Double-White-Dwarf Mergers and Connections Between Extreme Helium Stars, R CrB stars, Hot Subdwarfs, and,
Possibly, Type Ia Supernovaemostly with Karakas, Zhang, Yu and Saio
Friday, 20 April 12
Outline
• Double-White-Dwarf Binaries
• Double-White-Dwarf Mergers
• Hydrogen-Deficient Stars
• CO+He mergers
• He+He Mergers [Xianfei Zhang]
• Some Statistics
• Other DD Merger outcomes
Friday, 20 April 12
Double White Dwarf Binaries
• Binary star evolution produces DWDs• Gravitational-wave radiation implies orbit shrinks
• What happens next? –i) violent interaction–ii) stable interaction
• Keys:–stability, efficiency, progenitors
• Can we identify the products?Friday, 20 April 12
Low-mass DWDsMany new DWD binaries found (Kilic et al. 2011, Marsh et al. , Brown et al., ...). Discrepancies about stability. Predicted mass-period distribution sensitive to physics (eg WD cooling). All agree many short-period systems will merge. What happens next?
5
Friday, 20 April 12
LISA visible
Gra
vita
tiona
l wav
e st
reng
th
Orbital Period (log s)
LISA invisible
Post CE-system
Gravitational wave evolution
Friday, 20 April 12
Reflections on stability: progenitor stats
Nelemans 2001
Direct impact accretion
Violent accretion
HE+HE
CO+HE
AM CVn
q=1 q=2/3
Friday, 20 April 12
Reflections on stability: pyrotechnics
What happens when accretion rates< “violent” ?Piersanti et al. 2011, IAU Symp 281
9
Friday, 20 April 12
Reflections on stability: synchronization
Marsh et al. 2004 (but see talk by Sepinsky)
q=1 q=2/3
Friday, 20 April 12
Reflections on efficiency
Han & Webbink 1999Assume spherical symmetry and require
Marcello’s talk:Hydro simulation non-spherical β≈1 for q=0.4
M < MEdd
q=0.4β=0.7➝
q=2/3
Friday, 20 April 12
Yoon et al. 2007. Also: Benz et al. 1990ab, Segretain et al. 1997, Guerrero et al. 2004, Loren-Aguilar et al. 2009, Marcello et al. 2012...
evolution of a 0.9+0.6 M CO WD
Friday, 20 April 12
Groot 2012: Post-Merger systems
What are they?Single massive WD, R CorBor, sdB, NS, low-mass BH?
Where are they?Are they lurking in our knowledge/archives?
How do we recognize them? How do we tie them to their DD past?
Friday, 20 April 12
Potential Products
• He+He ⇒ He ignition ⇒ sdO/B star ⇒ He/CO WD (Nomoto & Sugimoto 1977, Nomoto & Hashimoto 1987, Kawai, Saio & Nomoto 1987, 1988, Iben 1990, Saio & Jeffery 2000, Zhang & Jeffery 2012)
• ?? He/CO+He ⇒ He ignition ⇒ sdO star ⇒ CO WD (Justham et al. 2010)
• He+CO ⇒ RCrB / EHe star ⇒ CO WD OR explosion ? (Webbink 1984, Iben & Tutukov 1984, Iben 1990, Saio & Jeffery 2002)(Wang et al. 2010)
• CO+CO ⇒ C ignition ⇒ ONe WD OR collapse/explosion ?(Hachisu et al. 1986a,b, Kawai, Saio & Nomoto 1987, 1988, Nomoto & Hashimoto 1987, Mochkovitch & Livio 1990, Saio & Nomoto 1998)
• ONe+CO ⇒ collapse/explosion ?
• Need proper calculations of merger• Need proper calculations of evolution
– Results sensitive to WD temperature AND accretion rate
Friday, 20 April 12
DWD mergers are unlikely to be hydrogen-rich !
RCB
HdC
EHe
HesdB
HesdO-
HesdO+
O(He)
PG1159
DB DQDO
[WC-L]
[WC-E]
Friday, 20 April 12
R Coronae Borealis variables• ~ 35 known in galaxy,
17 in the LMC (Clayton’s web page)
• Irregular light fades (5m)• Low-amplitude pulsations• Hydrogen-deficient spectrum• Infrared excess
R CrBR CrB
90 in MACC ??
Friday, 20 April 12
Extreme Helium stars• Approx. 17 known in galaxy• Spectrum: A- and B-
• Strong HeI • Narrow lines: supergiant• No Balmer lines• Strong N and C
• Origin? - clues from• distribution• chemical composition• low-amplitude pulsations
Comparison of spectrum of an extreme helium star with a helium-rich B star.
Jaschek & Jaschek, 1987, The classification of stars, Cambridge
Friday, 20 April 12
Helium-rich subdwarfsPalomar-Green survey of faint-blue objects finds many hot subdwarfs.
1) sdB: He<1% > 2000 analyzed
2) sdO, sdOB, sdOC = He-sdO, sdOD = He-sdB
+SDSS+EC+HE: > 170 He-sds.
3) Spectroscopic analysis ➝extreme: H<10%intermediate: 10%<H<99%
Friday, 20 April 12
EHEs as merger products?
DWD Merger products will be H-deficient
CO+He → RCrB → EHe → HesdO+ → O(He)??
Friday, 20 April 12
helium ignites in shell at core-envelope boundary
helium-burning shell forces star to expand to yellow giant, ~103 yr
accretion turned off at selected final mass
0.6 M CO-WD accretes He at 10-5 M /yr
0.5 M CO-WD
0.6 M , X=0.001
CO+He WD mergers
Saio & Jeffery, 2002, MNRAS
hypothesis
CO+He WD binary formed
orbit decays
less massive WD disrupted when Porb ~4 minutes
forms thick disk?
CO WD accretes material from disk
⇒model
Friday, 20 April 12
CO+He merger: EHes and RCrBs
CO+He mergerssolid: 0.6MCO+ x M Hedashed: 0.5MCO+ x M He
light: accretionheavy: contraction
EHe stars
RCrB stars
CO+He mergers solid: 0.6 M CO + x M Hedashed: 0.5 M CO + x M He
He+He mergerdotted: 0.7 M He+He
(Saio & Jeffery, 2002, MNRAS)Friday, 20 April 12
Extreme Helium Star and RCrB star surface abundances from spectroscopic analyses.
Each panel shows abundance of one element relative to iron, where 0,0 represents the solar value.
Blue Squares: Extreme Helium StarsRed Diamonds: RCrB (majority)Red Triangles: RCrB (minority)
(Jeffery, Heber, Pandey, Asplund, Lambert, ...1993 - 2011)
Chemical Clues
Friday, 20 April 12
a) H (not shown) relicb) Ca,Ti,Cr,Mn,(Ni) ~ ∝ Fe OKc) [N/Fe] ∝ [(C+N+O)/Fe] CNOd) [C/Fe] >> 0, 12C >> 13C 3αe) [O/Fe] >> 0, 18O ≈ 16O 14N+α?f) [Ne/Fe] >> 0 14N+2αg) Mg,Si,S,… X + αh) [F/Fe]>>0 18O+p?i) [s/Fe] >> 0 AGB ?j) [P/Fe] >> 0 AGB ?k) Li present ??
Astroarchaeology:a star digs up its past.
Object: deduce previous evolution from present chemical composition of stellar surface
Friday, 20 April 12
Understanding EHe abundances: the simple recipe: Mk II
HHe
C/O He
H
He+C+HC/O
CO WD from AGB starCO core massHe intershell massHe intershell abundancesdepend on initial M and ZH mass negligible
He WD from post-MS star
He core abundancesfrom CNO burning
H mass negligible
Mcore ~ f(Mi,Z) (agb models)
Mshell ~ f(Mi,Z) (agb model)
Mmerger ~ (1 + qcrit ) * (Mcore+Mshell)
qcrit: defined by dr
dm>
da
dq
Friday, 20 April 12
The mass of the AGB progenitor has a strong influence on the composition of the intershell and, hence, possibly, on the surface composition of any subsequent WD merger.
What clues do surface abundances provide about previous evolution?
[N/Fe] ∝ [(C+N+O)/Fe] OK[C/Fe] >> 0 fixed[O/Fe] >> 0 18O pocket[Ne/Fe] >> 0 ??Mg,Si,S,… OK[F/Fe]>>0 Magb~2-3M ?[s/Fe] >> 0 not modelled[P/Fe] >> 0 Magb~2-3M ?Li ??So far, not so good , we still need: • s-process yields from the AGB grids• experiment with hot merger models
Simple CO+He merger models for AGB stars with initial masses:
1.0, 1.9, 3.0, and 5.0 M
Friday, 20 April 12
18O: Signature of a hot or cold merger?
Hot merger (Clayton et al. 2007), 18O from partial α-burning of 14N
Could produce extra 19F
Cold merger18O from pocket at CO/He boundary in CO WD -- needs mixing over boundary, but it works. Can give some 19F from AGB.
Friday, 20 April 12
Progenitor and final masses?Observations (pulsation,
distribution):MEHe ≈ 0.7 – 0.9 M
Thick Disk / Bulge
AGB evolution:19F, 31P: M1 ≈ 2 – 3 M
MCOWD ≈ 0.55 – 0.65 M
If MHeWD ≈ 0.25 – 0.3 M ⇒ Mmerger ≈ 0.85 – 1.0 M
(Jeffery et al. 2011)
Population Synthesis ??M1 ≈ 1.4 – 3 M
MHeWD ≈ 0.25 – 0.35 M
MCOWD ≈ 0.4 – 0.65 M
⇒ Mmerger ≈ 0.65 – 1.0 M
Thin Disk ?
Friday, 20 April 12
HesdBs as merger products ?
Merger products will be extremely H-deficientHe+He → EHe → HesdB → He-sdO/B
→ sdB? see talk by Xianfei Zhang
Friday, 20 April 12
Merger statistics • Do DWDs occur in the right period and
mass ranges to produce mergers?
• Do they come from old stellar populations or recent star formation?
• What are there progenitors?
• Do mergers occur at the right frequency to explain populations of “products”?
• Some preliminary findings.... (with Yu and Nelemans, ...)
Friday, 20 April 12
• Code: modified Hurley (Yu & Jeffery 2010)• Sample: 1,000,000 binaries• IMF: Kroupa 1993 (broken power law)• Mass transfer: αλ=1• Mass ratio: dN/dq = 1• Separation: dN/d log a = constant (Han 1998)• Eccentricity: dN/de = 2e• Z=0.02• Initial-final mass relation: Hurley (2002)• No star formation convolution
Population Synthesis
Friday, 20 April 12
0 2000 4000 6000 8000 1000012000140000510152025303540455055606570
Number distribution for DWDs as a function of time to merger (Myr )
0 2000 4000 6000 8000 10000 12000 140000
50
100
150
200
250
300
350
400
450
He+He: old environments (thick disk, bulge, halo) -- OKCO+He:Mco>0.6 young environments (thin disk) Mco<0.6 older environments (thick disk ?)
Friday, 20 April 12
-2 -1 0 1 2 3
0.3
0.4
0.5
0.6
0.7
0.8
0.9
02.0004.0006.0008.00010.0012.0014.0016.0018.0020.0022.0024.0026.0028.0030.00
Mass-ratio / period distribution for DWDs
-3 -2 -1 0 1 2 3
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.00
2.000
4.000
6.000
8.000
10.00
12.00
14.00
16.00
18.00
20.00
-3 -2 -1 0 1 2 3
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.00
2.000
4.000
6.000
8.000
10.00
12.00
14.00
16.00
18.00
20.00
-3 -2 -1 0 1 2 3
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.00
2.000
4.000
6.000
8.000
10.00
12.00
14.00
16.00
18.00
20.00
Friday, 20 April 12
Interim Results
• Most CO+He mergers from recent star formation (~1 Gyr) - contradicts galactic distribution of EHe+RCrB
• CO+He mergers from 1.5 - 3 M⦿ progenitors- supported by nucleosynthesis and abundances in EHe+RCrB
• Most high-mass He+He mergers from 4 - 8 Gyr
• Most low-mass He+He mergers from >8 Gyr- supported by thick disk location of sdBs
• He+He mergers from 1.0 - 1.5 M⦿ progenitors
Friday, 20 April 12
DD mergers and Mch
i) Include non-conservative mass transfer in merger
ii) Require M2/M1>1.5and M2>0.9(1.0) in order to avoid AIC (Pakmor et al. 2010,11,12)
SNIa rates reduced by factors between 2/3 and 1/6(Chen et al. in prep)
q=2/3
q=1
Z�M1(t)dt+M2 = Mch
�M1 +M2 = Mch
M1 +M2 = Mch
Friday, 20 April 12
DD merger outcomes: q>qcrit
He+HeHesdO/B stars? EHe stars
CO+HeRCrB + EHe stars? ONe WDs
CO+CO Mtot>Mch & M2 > 1.0: SN IaMtot>Mch & M2 < 0.9: AICMtot<Mch: sub Chandra Ia?...? neutron stars
q=2/3
CO+HeRCrBHe+He
HesdO
AIC.Ia
Ia
AM CVn
???
???.Ia ?CO+CO
???AIC
Friday, 20 April 12
Conclusions• EHe+RCB abundances consistent with CO+He
merger. Require some boundary layer mixing and a hot merger. Progenitors 2 - 3 M⦿
• He-sdO/B abundances consistent with He+He merger. Require a composite (corona+disk) process. The most massive mergers are C-rich.
• EHe+RCB distribution suggests an old population. Binary Pop Synth suggests CO+He DWDs primarily young (~1 Gyr). Progenitors 1.5 - 3 M⦿
• He+He DWDs generally much older (4-8 Gyr)Progenitors 1 - 1.5 M⦿
Friday, 20 April 12